Raquel Urtasun

105

Papers

3,151

Total Citations

Papers (105)

Understanding the Effective Receptive Field in Deep Convolutional Neural Networks

NeurIPS 2016arXiv

The Reversible Residual Network: Backpropagation Without Storing Activations

NeurIPS 2017arXiv

Few-Shot Learning Through an Information Retrieval Lens

NeurIPS 2017arXiv

Learning Deep Parsimonious Representations

Proximal Deep Structured Models

UnO: Unsupervised Occupancy Fields for Perception and Forecasting

Learning to Segment Under Various Forms of Weak Supervision

Holistic 3D Scene Understanding From a Single Geo-Tagged Image

segDeepM: Exploiting Segmentation and Context in Deep Neural Networks for Object Detection

Instance-Level Segmentation for Autonomous Driving With Deep Densely Connected MRFs

Monocular 3D Object Detection for Autonomous Driving

HD Maps: Fine-Grained Road Segmentation by Parsing Ground and Aerial Images

MovieQA: Understanding Stories in Movies Through Question-Answering

Efficient Deep Learning for Stereo Matching

Efficient Multiple Instance Metric Learning Using Weakly Supervised Data

Sports Field Localization via Deep Structured Models

Deep Watershed Transform for Instance Segmentation

Annotating Object Instances With a Polygon-RNN

GeoNet: Geometric Neural Network for Joint Depth and Surface Normal Estimation

Deep Parametric Continuous Convolutional Neural Networks

SurfConv: Bridging 3D and 2D Convolution for RGBD Images

Hierarchical Recurrent Attention Networks for Structured Online Maps

Fast and Furious: Real Time End-to-End 3D Detection, Tracking and Motion Forecasting With a Single Convolutional Net

PIXOR: Real-Time 3D Object Detection From Point Clouds

Matching Adversarial Networks

SBNet: Sparse Blocks Network for Fast Inference

Learning Deep Structured Active Contours End-to-End

Deep Rigid Instance Scene Flow

Multi-Task Multi-Sensor Fusion for 3D Object Detection

DARNet: Deep Active Ray Network for Building Segmentation

End-To-End Interpretable Neural Motion Planner

UPSNet: A Unified Panoptic Segmentation Network

Convolutional Recurrent Network for Road Boundary Extraction

Learning to Localize Through Compressed Binary Maps

OctSqueeze: Octree-Structured Entropy Model for LiDAR Compression

PolyTransform: Deep Polygon Transformer for Instance Segmentation

Physically Realizable Adversarial Examples for LiDAR Object Detection

LiDARsim: Realistic LiDAR Simulation by Leveraging the Real World

PnPNet: End-to-End Perception and Prediction With Tracking in the Loop

SceneGen: Learning To Generate Realistic Traffic Scenes

S3: Neural Shape, Skeleton, and Skinning Fields for 3D Human Modeling

Deep Multi-Task Learning for Joint Localization, Perception, and Prediction

GeoSim: Realistic Video Simulation via Geometry-Aware Composition for Self-Driving

TrafficSim: Learning To Simulate Realistic Multi-Agent Behaviors

MP3: A Unified Model To Map, Perceive, Predict and Plan

Permute, Quantize, and Fine-Tune: Efficient Compression of Neural Networks

AdvSim: Generating Safety-Critical Scenarios for Self-Driving Vehicles

Virtual Correspondence: Humans as a Cue for Extreme-View Geometry

Implicit Occupancy Flow Fields for Perception and Prediction in Self-Driving

UniSim: A Neural Closed-Loop Sensor Simulator

Towards Unsupervised Object Detection From LiDAR Point Clouds

MixSim: A Hierarchical Framework for Mixed Reality Traffic Simulation

Learning Compact Representations for LiDAR Completion and Generation

Aligning Books and Movies: Towards Story-Like Visual Explanations by Watching Movies and Reading Books

Enhancing Road Maps by Parsing Aerial Images Around the World

Monocular Object Instance Segmentation and Depth Ordering With CNNs

Lost Shopping! Monocular Localization in Large Indoor Spaces

FollowMe: Efficient Online Min-Cost Flow Tracking With Bounded Memory and Computation

Be Your Own Prada: Fashion Synthesis With Structural Coherence

Towards Diverse and Natural Image Descriptions via a Conditional GAN

TorontoCity: Seeing the World With a Million Eyes

DeepRoadMapper: Extracting Road Topology From Aerial Images

SGN: Sequential Grouping Networks for Instance Segmentation

Situation Recognition With Graph Neural Networks

3D Graph Neural Networks for RGBD Semantic Segmentation

DAGMapper: Learning to Map by Discovering Lane Topology

DSIC: Deep Stereo Image Compression

DMM-Net: Differentiable Mask-Matching Network for Video Object Segmentation

DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatch

Learning Joint 2D-3D Representations for Depth Completion

Self-Supervised Representation Learning From Flow Equivariance

LookOut: Diverse Multi-Future Prediction and Planning for Self-Driving

Adversarial Attacks on Multi-Agent Communication

Towards Zero Domain Gap: A Comprehensive Study of Realistic LiDAR Simulation for Autonomy Testing

MemorySeg: Online LiDAR Semantic Segmentation with a Latent Memory

Real-Time Neural Rasterization for Large Scenes

Learning Lane Graph Representations for Motion Forecasting

GenAssets: Generating in-the-wild 3D Assets in Latent Space

Deep Feedback Inverse Problem Solver

Weakly-supervised 3D Shape Completion in the Wild

Conditional Entropy Coding for Efficient Video Compression

RadarNet: Exploiting Radar for Robust Perception of Dynamic Objects

DSDNet: Deep Structured self-Driving Network

Perceive, Predict, and Plan: Safe Motion Planning Through Interpretable Semantic Representations

Implicit Latent Variable Model for Scene-Consistent Motion Forecasting

Testing the Safety of Self-driving Vehicles by Simulating Perception and Prediction

Rethinking Closed-Loop Training for Autonomous Driving

3D Object Proposals for Accurate Object Class Detection

Skip-Thought Vectors

NeurIPS 2015arXiv

V2VNet: Vehicle-to-Vehicle Communication for Joint Perception and Prediction

MAD: Memory-Augmented Detection of 3D Objects

DIO: Decomposable Implicit 4D Occupancy-Flow World Model

Neuroaesthetics in Fashion: Modeling the Perception of Fashionability

Real-Time Coarse-to-Fine Topologically Preserving Segmentation

Rent3D: Floor-Plan Priors for Monocular Layout Estimation

Neural Guided Constraint Logic Programming for Program Synthesis

Efficient Graph Generation with Graph Recurrent Attention Networks

LoCo: Local Contrastive Representation Learning

MuSCLE: Multi Sweep Compression of LiDAR using Deep Entropy Models

Neural Lighting Simulation for Urban Scenes

Learning Deep Structured Models

Training Deep Neural Networks via Direct Loss Minimization

Deep Spectral Clustering Learning

Reviving and Improving Recurrent Back-Propagation

Learning to Reweight Examples for Robust Deep Learning